1
0
mirror of https://github.com/TomHarte/CLK.git synced 2024-11-19 08:31:11 +00:00
Commit Graph

1244 Commits

Author SHA1 Message Date
Thomas Harte
6a79ce9eb1 Adds enough to the Apple II's video that I can see what's going on with soft switches. 2018-04-15 21:55:26 -04:00
Thomas Harte
465c38f03c Extends the keyboard protocol and adds keyboard input to the Apple II. 2018-04-15 21:11:30 -04:00
Thomas Harte
be05d51e07 Now gives something a lot like the proper character output. 2018-04-15 20:31:04 -04:00
Thomas Harte
9bc470027e Put enough in place to get a visual representation of video memory.
Not the correct one, and so as to indicate that the machine isn't booting, surprisingly.
2018-04-15 19:35:08 -04:00
Thomas Harte
335c633884 Retrenches temporarily to full 8bpp output; introduces extra half a colour cycle of pause. 2018-04-15 18:54:05 -04:00
Thomas Harte
cd26f11818 Fixes documentation misstatement. 2018-04-15 18:00:51 -04:00
Thomas Harte
abe47b6ed8 Makes first attempt at a stable display area. Not entirely successful. 2018-04-15 18:00:40 -04:00
Thomas Harte
61659faeaa Adds the necessary call-outs to allow implementation of video generation. 2018-04-15 15:13:07 -04:00
Thomas Harte
71adb964e5 The Apple II now has a functioning processor, ROM and RAM. 2018-04-14 21:41:26 -04:00
Thomas Harte
e599e65087 Switches to use of the TargetList typedef wherever possible. 2018-04-14 19:46:38 -04:00
Thomas Harte
7efee9b52b Does the bare minimum to create a class skeleton for Apple II implementation. 2018-04-14 19:46:15 -04:00
Thomas Harte
079dc671e1 Rationalises per-machine static analyser call pattern, and adds Apple II as an option. 2018-04-14 12:12:12 -04:00
Thomas Harte
15f7cbe8c1 Corrects capitalisation. 2018-04-11 22:06:50 -04:00
Thomas Harte
428b6145fa Converts 6560 to more project normative templated form. 2018-04-11 22:00:42 -04:00
Thomas Harte
3ad0b31db8 Limits regions accessible to the 6560 to those built into the machine. 2018-04-11 21:35:23 -04:00
Thomas Harte
94b1c37fb2 Slightly simplifies bus decoding. 2018-04-08 18:51:37 -04:00
Thomas Harte
b20cbcd5fe Causes the Vic-20 to obey its own has_c1540 flag. 2018-04-08 17:35:02 -04:00
Thomas Harte
865c47a1ac Names the magic constants. 2018-04-08 10:35:07 -04:00
Thomas Harte
3821679efd Switches to bit unpacking on the GPU. 2018-04-07 22:17:47 -04:00
Thomas Harte
10f637d2cf Enhances the CompoundSource so that constituents can have different volumes. 2018-04-07 14:30:02 -04:00
Thomas Harte
36e0cb29c0 Ensures proper propagation of video choice through the Oric. 2018-04-04 19:14:42 -04:00
Thomas Harte
c0b4dd65da Mades the expected video signal usage explicit. 2018-04-04 19:01:18 -04:00
Thomas Harte
69f520428d Makes a first, ugly attempt at a 'new machine' dialogue for the Mac.
Which has implied getting much more specific about MSX disk drive attachment, and has prompted an excuse to offer the ZX80 with the ZX81 ROM.
2018-04-02 22:42:41 -04:00
Thomas Harte
33084899d0 Provides s-video as a command-line option. 2018-03-31 22:14:34 -04:00
Thomas Harte
8b45377b89 Simplifies storage underlying Vic memory.
In the hope of avoiding non-obvious bugs.
2018-03-31 18:54:40 -04:00
Thomas Harte
f6fb368d88 Allows the fast-tape mechanism to take ownership of tape handling.
Any successful fast tape interaction will now permanently pause the tape until a failed interaction occurs. This may or may not be a good idea.
2018-03-30 21:22:52 -04:00
Thomas Harte
912791d3d4 Causes the s-video path correctly to function. 2018-03-30 18:24:18 -04:00
Thomas Harte
163a61dd63 Corrects SVideo-as-composite output; the Atari and Vic-20 now both supply svideo. 2018-03-30 13:16:18 -04:00
Thomas Harte
33281b9d89 Introduces S-Video as a video signal type at the interface level. 2018-03-30 10:25:41 -04:00
Thomas Harte
286259c83b Adds missing 6560 update hooks. 2018-03-29 20:49:36 -04:00
Thomas Harte
ec353ce663 Makes minor Vic-20 optimisations.
Specifically: the 6560 is updated only upon writes (more nuance can arrive), and tape sleeps are observed.
2018-03-27 21:52:52 -04:00
Thomas Harte
6d464557a0 Reintroduces a warm-up run for the C1540.
That simulates the normal real-life scenario of switching the drive on slightly before the computer, and causes it to function correctly from immediate fast typing on an American Vic.

Also switches a couple of casts within the C1540 to functional style.
2018-03-26 21:06:07 -04:00
Thomas Harte
a2da51c30b Commutes Vic-20 machine configuration options to its Target. 2018-03-26 19:01:57 -04:00
Thomas Harte
e15d6717a1 Strips back Vic-20 static analysis to the bare minimum.
Also corrects an unsafe assumption in fast loading.
2018-03-25 13:37:33 -04:00
Thomas Harte
7fcd243be0 Ensures the rom fetcher is properly recorded for potential provision to the C1540. 2018-03-23 18:20:17 -04:00
Thomas Harte
a301964bd0 Ensures all audio queues are fully merged before machine destruction.
Thereby avoids a race condition.
2018-03-22 21:59:19 -04:00
Thomas Harte
2712702461 Makes get_clock_rate protected. It's now an implementation detail. 2018-03-22 10:01:18 -04:00
Thomas Harte
da3d65c18f Devolves time to cycle conversion to machines.
Thereby avoids a whole bunch of complicated machinations that would otherwise have been required of the multimachine.
2018-03-21 22:18:13 -04:00
Thomas Harte
89bec2919f Encapsulates machine configuration properties for all remaining platforms. 2018-03-09 16:07:29 -05:00
Thomas Harte
78eaecb29e Provides the proper framework for encapsulation of analyser target specifics.
... while making them a safe container for objects too. Uses the ZX80/81 as the pilot platform.
2018-03-09 15:36:11 -05:00
Thomas Harte
6b1eef572b Eliminates diamond inheritance of KeyboardMachine::Machine by typers.
Specifically by pulling the key action stuff into a purely abstract class [/interface]. Takes the opportunity to unpublish a bunch of machine details.
2018-03-09 15:19:02 -05:00
Thomas Harte
48737a32a7 Introduces formal setting of the output volume to SampleSource.
Previously every output device was making its own decision. Which is increasingly less sustainable due to the CompoundSource.
2018-03-09 13:23:18 -05:00
Thomas Harte
03501df9e5
Merge pull request #365 from TomHarte/CartridgeDetermination
Works towards eliminating the special cases for Atari 2600 ROM handling.
2018-03-08 18:40:58 -05:00
Thomas Harte
1804ea6849 Ensures the ColecoVision and MSX account for instruction lengths in advance when timing secondary components. 2018-03-07 17:00:18 -05:00
Thomas Harte
c8657e08f4 Merge remote-tracking branch 'origin/master' into CartridgeDetermination 2018-03-07 16:42:16 -05:00
Thomas Harte
9abc020818 Corrects potential ColecoVision SGM AY timing issues. 2018-03-07 16:16:58 -05:00
Thomas Harte
2dade8d353 Introduces ZonX emulation for the ZX81. 2018-03-07 16:16:29 -05:00
Thomas Harte
1100dc6993 Opens up .bin and .rom to all cartridge platforms, and adds a confidence estimate to the Atari 2600. 2018-03-07 14:26:07 -05:00
Thomas Harte
2452641844 Introduces a fast workaround to avert a MultiMachine where it would instantly end. 2018-03-06 19:08:02 -05:00
Thomas Harte
c82af4b814 Introduces get_confidence for the ColecoVision.
Based almost entirely on joypad accesses for now.
2018-03-06 19:06:35 -05:00
Thomas Harte
3c5a8d9ff3 Adds Super Game Module support for the ColecoVision. 2018-03-03 13:08:33 -05:00
Thomas Harte
bdfc36427c Implements MegaCart support. 2018-03-02 18:40:01 -05:00
Thomas Harte
59b4c7314d Merge branch 'master' into ColecoVision 2018-03-01 22:01:26 -05:00
Thomas Harte
b05d2b26bf Corrects a counting error in the MSX. 2018-03-01 21:59:51 -05:00
Thomas Harte
86239469e7 Allows SN76489 consumers to apply an additional divider that reduces computation. 2018-03-01 18:51:05 -05:00
Thomas Harte
7890506b16 Gives the SN76489 its proper dividers and personalities. 2018-02-28 22:36:03 -05:00
Thomas Harte
87760297fc Fixes underpumping of SN76489.
Audio works now. Though I still need properly to confirm who owns dividers in practice. I think probably all division should be within the SN.
2018-02-27 22:59:29 -05:00
Thomas Harte
d4df101ab6 Makes a first attempt at implementing the SN76489. 2018-02-27 22:25:12 -05:00
Thomas Harte
0ad2676640 Adds a class for the SN76489 and wires it into the ColecoVision. 2018-02-26 22:04:34 -05:00
Thomas Harte
a074ee2071 Possibly fixes ColecoVision input mapping.
Also provides symbolic input from the Mac.
2018-02-25 22:47:47 -05:00
Thomas Harte
204d5cc964 Extends JoystickMachine protocol to cover ColecoVision use case.
Also thereby implements input on the ColecoVision, in theory at least. No input is being fed though, so...
2018-02-25 19:08:50 -05:00
Thomas Harte
23d15a4d6c The ColecoVision now accepts and loads cartridges. 2018-02-24 18:26:44 -05:00
Thomas Harte
23c47e21de Proceeds the ColecoVision to booting. 2018-02-24 18:14:38 -05:00
Thomas Harte
5530b96446 Wired up a class and analyser for a ColecoVision. 2018-02-23 22:47:15 -05:00
Thomas Harte
270723ae72 Forces the MSX's perform_machine_cycle into the Z80. 2018-02-19 19:54:42 -05:00
Thomas Harte
fc81bfa59b Eliminates tape player call when tape is not playing. 2018-02-19 18:36:31 -05:00
Thomas Harte
3673cfe9be Pulls method call for tape fast loading checks out of inner loop for the Vic, Electron and ZX80/81. 2018-02-19 16:57:24 -05:00
Thomas Harte
b0ab617393 Simplifies inner loop test for MSX fast loading. 2018-02-19 16:24:47 -05:00
Thomas Harte
7b420d56e3 Removed state mirroring in the machine-specific Mac UI classes. 2018-02-14 21:46:50 -05:00
Thomas Harte
ddf1bf3cbf Reintroduces options selection for the Mac.
For everything except the Vic-20, anyway. That has a somewhat outdated notion of what an options panel should be, corresponding to the work yet to do on its analyser.
2018-02-12 21:46:21 -05:00
Thomas Harte
23105956d6 Fixes spurious unrecognised miss detection for the ASCII mappers. 2018-02-11 20:51:39 -05:00
Thomas Harte
f02989649c Corrects effect of pc_is_outside_bios. 2018-02-11 20:32:45 -05:00
Thomas Harte
eb39617ad0 Allows cartridges to filter based on the actor talking to them; corrects outstanding_machines access error. 2018-02-10 17:11:16 -05:00
Thomas Harte
43b682a5af Adds multiple target versions of all the DynamicMachine-vended types. 2018-02-09 16:31:05 -05:00
Thomas Harte
043fd5d404 Merge branch 'DynamicAnalysis' of github.com:TomHarte/CLK into DynamicAnalysis 2018-02-09 09:12:05 -05:00
Thomas Harte
d63a95983d Adds a couple of hard-stop conditions to the MSX, and respect for hard stops. 2018-02-09 09:10:56 -05:00
Thomas Harte
4cf258f952 Parallelises MultiMachine running, and ensures errors propagate. 2018-02-08 20:33:57 -05:00
Thomas Harte
4e720d57b2 With debugging hooks still on display, makes first attempt at dynamic analysis. 2018-02-01 07:53:52 -05:00
Thomas Harte
c12aaea747 Attempts to get as far as running the MultiMachine.
In doing so, fixes the long-standing bug that machines that output audio but don't have a listener produce a divide by zero.
2018-01-30 22:23:06 -05:00
Thomas Harte
ca48497e87 Pulls DynamicMachine out of MachineForTarget and adds MultiConfigurationTarget as a first multiplexer. 2018-01-29 21:49:49 -05:00
Thomas Harte
d493ea4bca Introduces a multimachine to handle multi-target static analyser outputs.
Non-functional as of yet.
2018-01-28 22:22:21 -05:00
Thomas Harte
e025674eb2 The MSX analyser is now smart enough not to be definitive when it's uncertain.
The cartridge type has also migrated to being a property of the cartridge, prefiguring my intention to discard the static analyser union.
2018-01-25 22:16:46 -05:00
Thomas Harte
f2519f4fd7 Decided to focus on 'confidence' over 'probability'.
Besides anything else, it individualises the measure. E.g. two targets can each have a confidence of 0.8 without each giving the wrong answer about probability.
2018-01-25 19:02:16 -05:00
Thomas Harte
66faed4008 Gives MachineForTargets complete responsibility for initial machine state. 2018-01-25 18:28:19 -05:00
Thomas Harte
11abc99ef8 Introduces the extra level of indirection necessary to make Analyser::Static::Target polymorphic. 2018-01-24 22:35:54 -05:00
Thomas Harte
21efb32b6f Integrates the static and nascent dynamic analyser namespaces. 2018-01-24 21:48:44 -05:00
Thomas Harte
d360b2c62d Standardises the static analyser on std::vector and slightly widens passageway to a machine.
The SDL target would now be fooled by a hypothetical multi-target, the Mac not yet.
2018-01-23 22:18:16 -05:00
Thomas Harte
6a112edc18 Corrects 16kb ASCII mapper.
Also increases hit position acceptance for the 8kb ASCII.
2018-01-22 22:13:16 -05:00
Thomas Harte
8fb4409ebb Adds hasty attempt at dynamic analysis to the MSX ROM handlers.
Logging for now, for further experimentation.
2018-01-22 21:50:56 -05:00
Thomas Harte
765c0d4ff8 Allows the Electron to load 8kb ROMs. 2018-01-15 21:27:45 -05:00
Thomas Harte
a5ac8c824e Removes logging and unnecessary get_drive_is_ready. 2018-01-07 21:59:59 -05:00
Thomas Harte
8be6cb827b Implements MSX interrupt/data request reading register.
The disk ROM now appears to accept on-disk bytes, but still announces an IO failure.
2018-01-07 20:28:34 -05:00
Thomas Harte
2f59226300 Fixes: DiskROM drive motor control, track_for_sectors' sides. 2018-01-07 20:02:40 -05:00
Thomas Harte
513c067f94 Makes an attempt to rope in the WD1770 for MSX disk ROM emulation. 2018-01-07 19:12:52 -05:00
Thomas Harte
655b971976 Establishes that there is such as a thing as a Konami SCC.
Creates one, ensures it appears in memory when intended to, lets it handle reads and writes. It currently does nothing.
2018-01-06 20:15:55 -05:00
Thomas Harte
185cd3c123 Expands and documents MSX::MemoryMap and MSX::ROMSlotHandler.
Hopefully to cover all intended use cases.
2018-01-06 15:51:29 -05:00
Thomas Harte
ed564cb810 Implements the main four cartridge banking schemes.
Slightly proof of concept for now.
2018-01-04 22:18:18 -05:00
Thomas Harte
c7f8f37822 Ensures typers terminate. 2017-12-30 13:46:30 -05:00
Thomas Harte
769d9dfbb9 Adds missing header. 2017-12-29 18:41:26 -05:00
Thomas Harte
6da8a3e24b Causes the MSX to respond to the appropriate standard configuration options. 2017-12-29 18:36:42 -05:00
Thomas Harte
e349161a53 Rejigs the typing relationship so that use of a typer is not strongly implied by the interface.
Simultaneously implements typing on the MSX by direct insertion into the key buffer.
2017-12-29 18:30:46 -05:00
Thomas Harte
d5b1a9d918 Moves the typer functionality behind a functionality-based naming scheme, eliminates its C-style memory management. 2017-12-29 15:26:03 -05:00
Thomas Harte
76af0228dd Corrects longstanding survival of camel case in the analyser's loadingCommand. 2017-12-29 15:15:29 -05:00
Thomas Harte
5fd0a2b9ea Attempts to pull reimplementations of TAPION and TAPIN better into line with originals.
Also improves whole flow of the fast tape hack that uses them.
2017-12-28 22:48:04 -05:00
Thomas Harte
11b73a9c0b Adds preliminary, non-error-checking wiring in of MSX parser alternatives to TAPION and TAPIN.
As both a prototype of the pending fast tape loading, and to provide for exact behaviour comparison.
2017-12-26 22:31:34 -05:00
Thomas Harte
b9b107ee85 Switches KeyGrave and KeyQuote, correcting a disarrangement. 2017-12-20 21:16:54 -05:00
Thomas Harte
0bb24075b6 Immediate fixes: TSX is seemingly TZX 1.21; the tape motor control works the other way around.
Input is not yet being recognised.
2017-12-19 22:17:42 -05:00
Thomas Harte
db6d9b59d0 Attempts to implement TSX support for the MSX. 2017-12-19 21:53:04 -05:00
Thomas Harte
2d892da225 Introduces the MSX keyboard toggle sample source.
In support of which, it also introduces a means of sample source composition.
2017-12-19 21:08:10 -05:00
Thomas Harte
d36e9d0b0d Reintroduces cstring.h to a few files that previously got it implicitly. 2017-12-18 21:47:30 -05:00
Thomas Harte
2dc1d4443e Separates LowpassFilter and SampleSource. 2017-12-18 21:39:23 -05:00
Thomas Harte
f8a2459c91 Corrects two lingering adaptation errors in the Vic-20. 2017-12-17 21:43:08 -05:00
Thomas Harte
ac80d10cd8 Separates the component parts of running an audio stream: task deferral, filtering and generation.
Walking towards improving opportunities for composition.
2017-12-17 21:26:06 -05:00
Thomas Harte
7a52e7d6d2 Provides an empty value for the interrupt cycle. 2017-12-13 22:44:03 -05:00
Thomas Harte
4daea1121b Gives up on C-BIOS for a while, to get to an acceptable merge point. 2017-12-12 21:19:33 -05:00
Thomas Harte
afcdd64d5e Switches to a less easy-to-confuse storage arrangement for MSX memory slots. 2017-12-11 21:09:53 -05:00
Thomas Harte
ec266d6c8e Ensures the AY stops listening to the bus after each read or write. 2017-12-04 19:18:54 -05:00
Thomas Harte
e3a5218e78 Fixes AY and random port input for the MSX. 2017-12-03 22:25:18 -05:00
Thomas Harte
3766bef962 Eliminates some redundant white space. 2017-12-03 14:52:42 -05:00
Thomas Harte
fe7fc6b22e Enables AY output from the MSX. 2017-12-02 16:30:43 -05:00
Thomas Harte
7f8a13a409 Adds bare minimum to get accepted 16- and 32kb cartridges to start on the MSX. 2017-12-02 16:06:04 -05:00
Thomas Harte
5da9cb2957 Introduces most of a keyboard mapping for the MSX. 2017-11-30 19:27:53 -05:00
Thomas Harte
54c845b6e2 Adds just enough logic to make every host key look like '0' to the MSX. 2017-11-29 22:07:30 -05:00
Thomas Harte
f0f149c018 Simplified paging logic. 2017-11-29 20:49:02 -05:00
Thomas Harte
aa4eef41d8 Seeks to introduce MSX interrupts. 2017-11-29 20:31:55 -05:00
Thomas Harte
69ec8a362e Makes an attempt to perform MSX memory paging. 2017-11-28 21:56:15 -05:00
Thomas Harte
d33612def5 Ensures the MSX provides a clock to the VDP. 2017-11-26 20:07:30 -05:00
Thomas Harte
9cb6ca3440 Adds elementary decoding of VDP accesses. 2017-11-26 20:01:11 -05:00
Thomas Harte
e957e40b14 Shifts 8255 logging up into its own port handler. That's probably fine for now. 2017-11-26 18:59:29 -05:00
Thomas Harte
7a8a43a96a Adds just enough of the MSX memory map for the Z80 to appear to try to do useful things. 2017-11-26 18:34:40 -05:00
Thomas Harte
0eb5dd9688 Introduces the fundamentals of bus routing for the MSX. 2017-11-26 16:47:59 -05:00
Thomas Harte
a14b53a9ab Adds a TMS9918 skeleton plus enough in the MSX to get to a blank screen in SDL/kiosk mode. 2017-11-26 13:28:26 -05:00
Thomas Harte
576d554a2c Expands upon the MSX skeleton. 2017-11-25 13:33:51 -05:00
Thomas Harte
68a2895753 Adds enough static analyser to get to the MSX itself as the point of failure in SDL/kiosk mode. 2017-11-25 13:18:24 -05:00
Thomas Harte
f90b3f06aa Merge branch 'master' into MSX 2017-11-25 08:19:24 -05:00
Thomas Harte
ee9f89ccb5 Simplifies CPC ROM input mechanism. 2017-11-25 08:18:01 -05:00
Thomas Harte
a46a37fba9 Ensures all vectors loaded from disk are the expected size. 2017-11-24 22:22:32 -05:00
Thomas Harte
324b57c054 Adds inclusion of the 3/4 of the MSX's support chips that are currently implemented. 2017-11-24 22:05:50 -05:00
Thomas Harte
ae50ca9ab2 Moves the MSX class to the appropriate place and gives it a Z80. 2017-11-24 21:59:54 -05:00
Thomas Harte
6e4bde00d3 Merge branch 'master' into MSX 2017-11-24 21:50:38 -05:00
Thomas Harte
221c05ca76 Adapts the Mac port to use a Machine::DynamicMachine, thereby eliminating plenty of duplication. 2017-11-24 21:36:22 -05:00
Thomas Harte
ff21ff90eb Introduces MSX ROMs and an MSX class. 2017-11-24 20:43:26 -05:00
Thomas Harte
c83b3cefbc Eliminates the generalised special case selectors and ROM suppliers from the CPC, Vic-20, Electron and ZX80/81. 2017-11-24 17:55:28 -05:00
Thomas Harte
a8ac51da73 Eliminates the Oric's non-reflective inputs for selections, and the Oric-specific ROM setter. 2017-11-24 16:59:00 -05:00
Thomas Harte
d6192b8c58 Ensures Oric video output starts up and changes validly. 2017-11-23 16:19:41 -05:00
Thomas Harte
c7dd6247f0 Corrects Electron Mode 3 timing. 2017-11-23 16:06:05 -05:00
Thomas Harte
3365ff0200 Adds type recipient as a dynamic type, and accepts paste and fullscreen toggle in SDL. 2017-11-21 21:44:29 -05:00
Thomas Harte
c6306db47c Ensures the 6560 is fully initialised by setup_output. 2017-11-21 21:24:06 -05:00
Thomas Harte
7b25b03cd5 Formally standardises machine options and introduces a --help option for the SDL target. 2017-11-20 21:55:32 -05:00
Thomas Harte
90d33949f9 Adds a mapping of backspace for the Electron. 2017-11-18 20:02:04 -05:00